The effects of selenium as sodium selenite fed at levels below the dietary requirement, near the requirement, and considerably above the requirement are being evaluated on the pancreatic cancer model induced by N-nitroso bis(2-oxypropyl)amine (BOP) in the Syrian hamster. Initially the three levels were selected by feeding hamsters excessively high levels of sodium selenite (0.25 up to 80.0 ppm Se) in a three week study and excessively low levels (0.0 up to 5.0 ppm Se) in a 25 week study. The selenium status of all animals was evaluated by growth; food consumption; blood and tissue selenium content; blood glutathione and glutathione peroxidase activity; and histopathology. In chronic studies with BOP the high and low levels of selenium are being fed to groups of hamster either only before, only after or both before and after a single injection of BOP with the control level, or the other extreme level fed during the remaining period. In addition rats, which develop coleorectal tumors when treated weekly with BOP, are being fed control or high levels of selenium to observe the influence of selenium on this tumor model with BOP. In both chronic studies blood will be collected at several intervals for the determination of blood selenium, glutathione peroxidase and glutathione concentrations. In addition to evaluating pancreatic histology, numerous other tissues will be evaluated in both BOP treated and their respective saline treated controls to evaluate the long term effect of feeding various levels of selenium on the longevity and general pathology of the Syrian golden hamster.